Software Selection and Configuration in Mobile Environments: A Utility-Based Approach
|
|
- Letitia Flynn
- 6 years ago
- Views:
Transcription
1 Software Selection and Configuration in Mobile Environments: A Utility-Based Approach Vahe Poladian Vahe.Poladian@cs.cmu.edu David Garlan Garlan@cs.cmu.edu Mary Shaw Mary.Shaw@cs.cmu.edu ABSTRACT Users of low-power mobile computing platforms make ad hoc decisions when choosing software components among alternatives and configuring those components. We propose applying utility-theoretic models, which can help determine optimal allocation of scarce resources to applications given the user s utility and application resource usage. We believe that taking into consideration resource consumption and applying microeconomic models has the potential of improving the user s satisfaction with the system. In this paper, we formulate the problem, demonstrate the use of a microeconomics-based model on a simple version of the problem, and list possible solutions. Further, we identify issues typical of mobile environments that are not addressed by existing research, and propose ways of tackling these issues. 1. MOTIVATION Mobile computer users are often faced with choosing among software components that provide similar services at various levels of quality (e.g., which map rendering program to use) and configuring those components (e.g., what portion of the map database to download). When making a selection among alternatives, users usually consider the computational features supported by similar applications and perhaps the dollar cost, but they typically ignore the differences among the resource requirements of the applications (e.g., a feature-rich application is likely to use more memory than a light-weight version). Mobile computers generally have limited resources (e.g., memory, disk bandwidth, battery capacity) as compared to desktop computers, so ignoring these resources can lead to substantially less useful performance than might be achieved. Most users typically do not apply a systematic method to Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Fourth International Workshop on Economics-Driven Software Engineering Research, May 21, 2002, Orlando, Florida. the configuration decision, and they often ignore, even informally, the burden that an application places on the resources of the mobile computer. The resulting ad hoc decisions might differentiate relatively good solutions from bad ones, but we believe that only a systematic approach will consistently yield optimal or even very good solutions. Thus we explore the potential for utility-theoretic models to be used in the problem of mobile software selection and configuration. 2. THE PROBLEM Let s formulate the problem. Given: A mobile computer system, equipped with scarce resources such as processing power, memory, network bandwidth, battery capacity, Configurable, fidelity-aware applications, which are capable of providing varying levels of quality in several dimensions of service (e.g., the Map Renderer might have the option to configure the size of its downloadable database, providing more detail and larger geographical coverage when more data is downloaded); and a choice between a unified application providing several services and several leaner, more specialized applications, Application profiles, which provide information about the various levels of quality in each service that an application provides, For each configuration of every application, resource usage information (e.g., the Map Renderer profile provides information about how much memory is needed in order to look up a certain number of locations), The preferences (utility) of a user with respect to all levels of quality for every desired service. 1 1 We assume that a mobile computer has only one user. Some of the related research has considered problems of mediating preferences of multiples users.
2 Determine: a configuration of the applications that provides the best overall utility to the user. What we mean by fidelity-awareness is the ability of the application to provide varying degrees of quality and resource usage trade-offs. Furthermore, we expect an application to report the quality levels that it provides, as well as the resource requirements for each quality level. We assume that the application will conform to its advertised resource usage, i.e., it will not consume more resources than it should for each level of quality. The space of quality levels, and hence the corresponding resource usage, can be either fine-grain (e.g., the granularity of the downloadable map database can be in bytes, and with each additional one hundred bytes, the number of locations that can be looked up increases by one), or coarse-grain (e.g., quality levels of low, medium, high of the application features, where low allows no attachments, medium selectively allows some attachments, and high allows all attachments to be downloaded). It is possible that various degrees of quality of essentially the same service are provided by more than one application. For example, a map application from one vendor might have a larger database, thus supporting better detail, as compared to a similar application from another vendor. In this case, several quality vs. resource trade-offs are available by the virtue of having different applications, hence choosing the best application amounts to selecting the best quality vs. resource trade-off. We can see that application configuration and application selection problems are intimately related. 3. RELATED WORK The Amaranth project [1], [5], [8] at Carnegie Mellon University has developed a Quality of Service based Resource Allocation Model (Q-RAM), which enables a computer system to optimally allocate its resources to maximize the system s utility. Our objective and approach are similar, but we address the mobile computing environment. Later we consider issues specific to mobile computers that we believe are not solved by previous research. The mobile systems group [3], [6] at University College London uses closed-bid Vickrey auctions to solve the problem of allocating bandwidth and other resources to multiple competing users in a mobile environment. This approach aims to mediate the preferences of multiple users. The Nemesis operating system [7] uses a decentralized approach in determining the optimal resource allocation. Processes reveal their preference for resources to the kernel. Using shadow prices, the system charges the consumer of a resource a price, which is the cost that other potential consumers of the resource incur by forgoing that resource. These charges are viewed as continuous feedback signals that allow the system to competitively adapt to a state that is optimal from the point of view of all consumers. Under the umbrella of Project Aura [1], [9] at Carnegie Mellon University, various groups investigate issues related to capturing user intent and providing for a distraction-free computing environment. In the architecture of Aura, the Environment Manager is the component that coordinates applications and configures them for use. We propose implementing utility theoretic models within Aura s Environment Manager. 4. APPROACH We are exploring the use of a microeconomic model that leverages knowledge of user s preferences, application quality of service vs. resource trade-offs, and resource constraints, to compute an optimal allocation of scarce computing resources among competing applications. The objective is to configure applications so as to maximize the user s utility. Using an example, we show the economic intuition behind the model. Consider a rather simplified model of computation, consisting of two simple applications: a map rendering application and an client, which can work in a disconnected mode. Assume, for simplicity, that the only resources under consideration are the battery of the device and its memory, e.g., flash memory. The Map Renderer provides directions and maps using a variable size downloadable database. The user can download only a portion of the entire database at a time. The bigger the database, the larger the region covered, and on average, the more locations that can be looked up. The application uses local memory to store messages for offline processing. The larger the amount of memory allocated to , the more messages that can be downloaded, read, and composed. The total memory of the mobile computing device imposes one constraint. The second constraint is imposed by the battery capacity, which is roughly proportional to the length of the time that the device can be used without re-charging. Naturally, the limited lifetime of the battery has to be allotted between the two applications. There are two configuration decisions to make: (1) how much memory to allocate to each application, and (2) how much energy (and thus, how much time) to allocate to using each application. For simplicity, we ignore the different policies that each application might use internally to manage memory and battery life. Treat the Map Renderer and applications as producers of map and services. As the unit of service provided by the Map Renderer application we consider the number of physical locations that are looked up using the application. Looking up one location requires little memory
3 and perhaps a few minutes of battery life. Looking up twenty locations requires proportionally more memory and longer battery usage. Similar analysis applies to the application, but with respect to the number s that are read, responded to, or composed. 2 The graph in Figure 1 introduces a visualization of the problem and the solution. The vertical axis shows the units of map service, and the horizontal axis, the units of service. Ignoring for a moment the battery constraint, we identify all the possible combinations of units of map and e- mail service that can be achieved given the memory constraint. Because of the linearity assumption, this constraint is a straight line, describing all the linear combinations of units of map and service that use up at most the total memory available. Similarly, we express the battery constraint (independent of the memory constraint) as another line, which can possibly have a different slope and intercepts. Notice that when the maximum number of map lookups is done, the memory of the device is entirely used up by the map data, but the battery is not yet drained. Similarly, when the maximum number of is processed, the battery is exhausted, but there is available memory. Thus that the map application is more memory-intensive, and the application is more processor-intensive. The region that is bounded above by the two bold lines shows all the possible combinations of map lookups and e- mails processed under the memory and battery constraints. But which is the optimal point? In order to determine that point, and the optimal resource allocation, we need to consider the utility of the user. As is common in microeconomics, let us consider the contours of equal utility, also called the indifference curves. Each indifference curve is made up of points that from the point of view of the user yield equal utility. The higher the curve, the better the utility. Typical shapes of indifference curves are described by equations such as M α x E (1-α) = const, where α is a real number between 0 and 1. In the graph we have one set of such curves, which indicate some utility bias towards the application. Although it may seem that the optimal solution should lie at the intersection of the resource constraint lines where both resources are saturated in fact it does not. In this case, the optimal solution occurs at a point where one of the indifference curves is tangent to the energy constraint line. Indeed, those curves that are higher (further away from the origin) are outside of the resource boundaries. And curves that are lower yield lower utility. Number of Map Lookups Done Energy Constraint Line Contours of Equal Utility (Indifference Curves) Optimal Point Memory Constraint Line Number of s processed Figure 1: The Optimal Point Yields the Highest Utility While Satisfying the Resource Constraints 2 We assume, for simplicity, that the level of service is linear in terms of each resource.
4 Economists use the concept of marginal utility to analyze the properties of an optimal solution. Marginal utility of e- mail service with respect to a resource is the incremental utility to the user, when a small amount of that resource is shifted to the application. Notice that at the optimal point, the marginal utility of service with respect to energy is equal to that of the map service (indicating that the indifference curve is tangent to the energy constraint line). This insight helps in generalizing the solution. This approach is also applicable in selecting one application among several alternatives. Simply treat the different providers of the same service as if they were different quality of service vs. resource trade-off opportunities provided by a single application. Apply the same analysis to determine the optimal point. The application that offers the latter combination is the best alternative. 5. SOLUTIONS TO THE PROBLEM The example just discussed makes several assumptions that may not be realistic in a computing environment, e.g., the quality space is continuous, that the mapping of quality space to the resource space is linear, that the utility is of particular functional form. These assumptions help us in understanding the economic intuition, but may need to be dropped in order to be applicable in the computer systems. The general problem described in section 2 is similar to the example we discussed, but allows for a larger number of applications and resources. Also, in the general case we drop many of the simplifying assumptions of the example. Classical economics solves the utility maximization problem using calculus. Assuming that the utility functions with respect to service level, and service level with respect to resource are all continuously differentiable and monotonic, then the optimal point has the property that the partial derivatives of the utility of each service with respect to a given resource are either all pair-wise equal, or some of those derivatives are pair-wise equal, and the rest are all zero 3. In computer systems, it is not realistic to assume that the quality space is continuous, and it may not be practical to make any assumptions about the properties of either quality or utility functions. For example, more resource may not consistently yield higher quality, and higher quality may not yield higher utility. However, we believe that there is value in studying the forms of utility functions that occur in mobile applications, as this can offer valuable insight to understanding the preferences of mobile users, and help in 3 Consider the change in total utility when shifting a slight amount of a resource from one service to another. If there are corner cases, then we can assume functions are piece-wise differentiable; however, the corner points need to be checked separately in case they are optimal. solving the optimization problem (e.g., as we showed above the optimal solution may not be the point where all the resources are saturated). Continuous models may provide reasonable approximations of fine-grained quality spaces. Lee [5] and Khan [4], in their approaches to the quality of service management, assume that the space of quality levels is a finite (discrete) set. Such an assumption is realistic, as applications typically offer only a finite set of quality vs. resource trade-offs. Given this assumption, the utility maximization problem can be formulated as a variation of a multi-dimensional knapsack problem (another common name for this set of problems is integer programming). It is known that these problems are NP-hard. The optimal solutions are exponential in the size of the input, but there are polynomial approximation algorithms that can find solutions within a threshold of the optimal. These models may provide reasonable approximations of coarse-grained quality spaces. 6. ISSUES SPECIFIC TO MOBILE COMPUTING ENVIRONMENTS We believe that mobile computing presents interesting problems that have not yet been addressed by previous research: (1) solving implementation issues specifically related to mobile systems, (2) describing the forms of utility functions that occur in mobile computing, and (3) dynamically reconfiguring the system after a change in the environment, resources, or utility. Let us discuss the latter group of issues in more detail, as it exposes several interesting research problems. Mobile computing environments change often. Such changes can be in resource availability (e.g., network bandwidth dropped, so perhaps a web image download should be deferred), in the environment (e.g., it is darker in one room than another, so the display needs to be brightened), in the utility of the user (e.g., after an urgent message regarding a shift in deadline, the utility of finishing the demo goes up, thus necessitating more resources for the demo). In response to all these changes, either the constraint or the utility or both changes, so the optimal point moves and the system needs to be reconfigured. We believe that current literature does not directly address this problem. The computation of a new optimal value may not be the best action in response to a change. To see this, imagine an increase in some resource that enables a better version of some application (e.g., that offers higher quality in some service) to run. Shutting down the currently running version and starting the better version incurs overhead cost hence that configuration may not yield higher utility than the current one. As another example, imagine a decrease in some resource, which necessitates that an application switch to a lower quality level. However, this might cause unnecessary user distraction, and may not be optimal from
5 the user s point of view. Accounting for hidden costs such as user distraction becomes an important problem in the reconfiguration, which was not an issue for the static version of the problem. Likewise, the cost of an incremental change to an existing configuration is not addressed by the simple model. Both the economic model described here and the solutions that we have found in the computing literature solve the static version of the problem. The aforementioned issues comprise only a small sub-set of all those that we have considered. We believe that it is possible to recast the problem of reconfiguration into utility models (i.e., to an integer programming problem) with the introduction of additional variables to account for overhead resource usage, costs of incremental change, hidden costs accounting for user distraction, etc. Such a reformulation might allow us to solve the problem of reconfiguration using existing solutions. However, existing solutions may not be computationally efficient enough for mobile computers. Ideally, we would like to reuse the computation of an optimal solution for an existing configuration in the process of computing the new optimum. This will make such solutions attractive for use in mobile environments, since computation is relatively expensive on mobile devices, and the changes are frequent. For some variations of integer programming there are solutions that are incremental. For example, in case of only one resource, there is a dynamic programming solution. We would like to explore the integer programming, linear programming, and other optimizations research literature for such solutions and algorithms that are well suited for incremental change. 7. CONCLUSION We have proposed using utility-theoretic models to determine user-optimal software configurations in mobile computing environments. Using an example, we have demonstrated how utility models can be applied to determine optimal solutions. We then formulated the generalized problem of mobile component selection and configuration, and mentioned approaches that promise to solve the problem. We also enumerated various issues that are characteristic of mobile environments that we believe are not addressed in current research. We propose approaches for tackling these issues, while reusing existing models. 8. ACKNOWLEDGMENTS This research is supported by the National Science Foundation under Grant CCR , by the Department of Defense under DARPA grants F and N , by the High Dependability Computing Program from NASA Ames cooperative agreement NCC , and by the Software Industry Center at Carnegie Mellon University. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the NSF, DOD, or the U.S. government. The authors would like to thank Joao Pedro Sousa, Orna Raz, and Dushyanth Narayanan of Carnegie Mellon School of Computer Science for critical feedback on this paper. 9. REFERENCES [1] The Amaranth Project at. [2] The Aura Project at. [3] L. Capra, W. Emmerich, and C. Mascolo. A Micro- Economic Approach to Conflict Resolution in Mobile Computing. UCL-CS Research Note RN/01/38. [4] S. Khan. Quality Adaptation in a Multisession Multimedia System: Model, Algorithms and Architecture. PhD Dissertation, Department of Electrical and Computer Engineering, University of Victoria, [5] C. Lee. On Quality of Service Management. PhD Dissertation, Department of Electrical and Computer Engineering,, [6] The Mobile Systems Group at the University College London. [7] R. Neugebauer and D. McAuley. Congestion Prices as Feedback Signals: An Approach to QoS Management. Proceedings of the 9th ACM SIGOPS European Workshop, September [8] R. Rajkumar, C. Lee, J. Lehoczky, and D. Siewiorek. A Resource Allocation Model for QoS Management. Proceedings of the IEEE Real-Time Systems Symposium, December [9] J.P. Sousa and D. Garlan. Aura: an Architectural Framework for User Mobility in Ubiquitous Computing Environments. Proceedings of the 3rd Working IEEE/IFIP Conference on Software Architecture, Montreal, August To appear.
ptop: A Process-level Power Profiling Tool
ptop: A Process-level Power Profiling Tool Thanh Do, Suhib Rawshdeh, and Weisong Shi Wayne State University {thanh, suhib, weisong}@wayne.edu ABSTRACT We solve the problem of estimating the amount of energy
More informationReservation Based Admission Control of a Multimedia Server
Reservation Based Admission Control of a Multimedia Server Md. Mujibur Rahman Dr. Mostofa Akbar Rukhsana Afroz Ruby Dept. of Computer Science Dept. of Computer Science & Engg, Millinium Solutions Ltd.
More informationAn Attempt to Identify Weakest and Strongest Queries
An Attempt to Identify Weakest and Strongest Queries K. L. Kwok Queens College, City University of NY 65-30 Kissena Boulevard Flushing, NY 11367, USA kwok@ir.cs.qc.edu ABSTRACT We explore some term statistics
More informationInternet Pricing. Abstract. 1 Introduction. 2 Interconnection. Md. Rafiqul Hasan Chowdhury Helsinki University of Technology
Internet Pricing Md. Rafiqul Hasan Chowdhury Helsinki University of Technology rafiqul.chowdhury@hut.fi Abstract Internet pricing can be seen from two points of view - between service providers and end-users
More information1.1 What is Microeconomics?
1.1 What is Microeconomics? Economics is the study of allocating limited resources to satisfy unlimited wants. Such a tension implies tradeoffs among competing goals. The analysis can be carried out at
More informationTHE GEOMETRY OF LINEAR PROGRAMMING
THE GEOMETRY OF LINEAR PROGRAMMING INTRODUCTION Understanding the relationship between geometry (or algebraic topology) and mathematical programming is critical to being able to apply the methods of mathematical
More informationSearch Costs vs. User Satisfaction on Mobile
Search Costs vs. User Satisfaction on Mobile Manisha Verma, Emine Yilmaz University College London mverma@cs.ucl.ac.uk, emine.yilmaz@ucl.ac.uk Abstract. Information seeking is an interactive process where
More informationGarbage Collection (2) Advanced Operating Systems Lecture 9
Garbage Collection (2) Advanced Operating Systems Lecture 9 Lecture Outline Garbage collection Generational algorithms Incremental algorithms Real-time garbage collection Practical factors 2 Object Lifetimes
More informationLecture 15: Segmentation (Edge Based, Hough Transform)
Lecture 15: Segmentation (Edge Based, Hough Transform) c Bryan S. Morse, Brigham Young University, 1998 000 Last modified on February 3, 000 at :00 PM Contents 15.1 Introduction..............................................
More informationAcknowledgement: BYU-Idaho Economics Department Faculty (Principal authors: Rick Hirschi, Ryan Johnson, Allan Walburger and David Barrus)
Math Review Acknowledgement: BYU-Idaho Economics Department Faculty (Principal authors: Rick Hirschi, Ryan Johnson, Allan Walburger and David Barrus) Section 1 - Graphing Data Graphs It is said that a
More informationGraphical Methods in Linear Programming
Appendix 2 Graphical Methods in Linear Programming We can use graphical methods to solve linear optimization problems involving two variables. When there are two variables in the problem, we can refer
More informationCapturing Design Expertise in Customized Software Architecture Design Environments
Capturing Design Expertise in Customized Software Architecture Design Environments Robert T. Monroe School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213 Abstract: Software architecture
More informationMapping Vector Codes to a Stream Processor (Imagine)
Mapping Vector Codes to a Stream Processor (Imagine) Mehdi Baradaran Tahoori and Paul Wang Lee {mtahoori,paulwlee}@stanford.edu Abstract: We examined some basic problems in mapping vector codes to stream
More informationAutomated Control for Elastic Storage
Automated Control for Elastic Storage Summarized by Matthew Jablonski George Mason University mjablons@gmu.edu October 26, 2015 Lim, H. C. and Babu, S. and Chase, J. S. (2010) Automated Control for Elastic
More informationDOWNLOAD PDF SYNTHESIZING LINEAR-ARRAY ALGORITHMS FROM NESTED FOR LOOP ALGORITHMS.
Chapter 1 : Zvi Kedem â Research Output â NYU Scholars Excerpt from Synthesizing Linear-Array Algorithms From Nested for Loop Algorithms We will study linear systolic arrays in this paper, as linear arrays
More informationMobile Offloading for Maximum Utility at Minimum Cost
ISSN 2278 0211 (Online) Mobile Offloading for Maximum Utility at Minimum Cost Anjana Antony M. Tech 4th Sem, Department of Computer Science & Engineering Dr. B. R. Prasad Babu Professor & HOD, Department
More informationPrices and Auctions in Markets with Complex Constraints
Conference on Frontiers of Economics and Computer Science Becker-Friedman Institute Prices and Auctions in Markets with Complex Constraints Paul Milgrom Stanford University & Auctionomics August 2016 1
More informationDistributed Scheduling of Recording Tasks with Interconnected Servers
Distributed Scheduling of Recording Tasks with Interconnected Servers Sergios Soursos 1, George D. Stamoulis 1 and Theodoros Bozios 2 1 Department of Informatics, Athens University of Economics and Business
More informationClassification of Optimization Problems and the Place of Calculus of Variations in it
Lecture 1 Classification of Optimization Problems and the Place of Calculus of Variations in it ME256 Indian Institute of Science G. K. Ananthasuresh Professor, Mechanical Engineering, Indian Institute
More informationSubject: Adhoc Networks
ISSUES IN AD HOC WIRELESS NETWORKS The major issues that affect the design, deployment, & performance of an ad hoc wireless network system are: Medium Access Scheme. Transport Layer Protocol. Routing.
More informationToward Low Cost Workload Distribution for Integrated Green Data Centers
Toward Low Cost Workload Distribution for Integrated Green Data Centers 215 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or
More informationReduction of Periodic Broadcast Resource Requirements with Proxy Caching
Reduction of Periodic Broadcast Resource Requirements with Proxy Caching Ewa Kusmierek and David H.C. Du Digital Technology Center and Department of Computer Science and Engineering University of Minnesota
More informationLinear Programming. Meaning of Linear Programming. Basic Terminology
Linear Programming Linear Programming (LP) is a versatile technique for assigning a fixed amount of resources among competing factors, in such a way that some objective is optimized and other defined conditions
More informationQuality-Assured Energy Balancing for Multi-hop Wireless Multimedia Networks via 2-D Channel Coding Rate Allocation
Quality-Assured Energy Balancing for Multi-hop Wireless Multimedia Networks via 2-D Channel Coding Rate Allocation Lin Xing, Wei Wang, Gensheng Zhang Electrical Engineering and Computer Science, South
More informationDOWNLOAD PDF BIG IDEAS MATH VERTICAL SHRINK OF A PARABOLA
Chapter 1 : BioMath: Transformation of Graphs Use the results in part (a) to identify the vertex of the parabola. c. Find a vertical line on your graph paper so that when you fold the paper, the left portion
More informationNetwork-Adaptive Video Coding and Transmission
Header for SPIE use Network-Adaptive Video Coding and Transmission Kay Sripanidkulchai and Tsuhan Chen Department of Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh, PA 15213
More informationCHAPTER 5 PROPAGATION DELAY
98 CHAPTER 5 PROPAGATION DELAY Underwater wireless sensor networks deployed of sensor nodes with sensing, forwarding and processing abilities that operate in underwater. In this environment brought challenges,
More informationA Secure and Dynamic Multi-keyword Ranked Search Scheme over Encrypted Cloud Data
An Efficient Privacy-Preserving Ranked Keyword Search Method Cloud data owners prefer to outsource documents in an encrypted form for the purpose of privacy preserving. Therefore it is essential to develop
More informationPhysically-Based Laser Simulation
Physically-Based Laser Simulation Greg Reshko Carnegie Mellon University reshko@cs.cmu.edu Dave Mowatt Carnegie Mellon University dmowatt@andrew.cmu.edu Abstract In this paper, we describe our work on
More informationGeneral properties of staircase and convex dual feasible functions
General properties of staircase and convex dual feasible functions JÜRGEN RIETZ, CLÁUDIO ALVES, J. M. VALÉRIO de CARVALHO Centro de Investigação Algoritmi da Universidade do Minho, Escola de Engenharia
More informationComplexity Measures for Map-Reduce, and Comparison to Parallel Computing
Complexity Measures for Map-Reduce, and Comparison to Parallel Computing Ashish Goel Stanford University and Twitter Kamesh Munagala Duke University and Twitter November 11, 2012 The programming paradigm
More informationHomework 2: Multi-unit combinatorial auctions (due Nov. 7 before class)
CPS 590.1 - Linear and integer programming Homework 2: Multi-unit combinatorial auctions (due Nov. 7 before class) Please read the rules for assignments on the course web page. Contact Vince (conitzer@cs.duke.edu)
More informationECE519 Advanced Operating Systems
IT 540 Operating Systems ECE519 Advanced Operating Systems Prof. Dr. Hasan Hüseyin BALIK (10 th Week) (Advanced) Operating Systems 10. Multiprocessor, Multicore and Real-Time Scheduling 10. Outline Multiprocessor
More informationAn Industrial Employee Development Application Protocol Using Wireless Sensor Networks
RESEARCH ARTICLE An Industrial Employee Development Application Protocol Using Wireless Sensor Networks 1 N.Roja Ramani, 2 A.Stenila 1,2 Asst.professor, Dept.of.Computer Application, Annai Vailankanni
More informationQuery Likelihood with Negative Query Generation
Query Likelihood with Negative Query Generation Yuanhua Lv Department of Computer Science University of Illinois at Urbana-Champaign Urbana, IL 61801 ylv2@uiuc.edu ChengXiang Zhai Department of Computer
More informationMinimum Bounding Boxes for Regular Cross-Polytopes
Minimum Bounding Boxes for Regular Cross-Polytopes Salman Shahid Michigan State University shahids1@cse.msu.edu Dr. Sakti Pramanik Michigan State University pramanik@cse.msu.edu Dr. Charles B. Owen Michigan
More informationCEMR position paper. on the recast of the proposal for a directive on waste electrical and electronic equipement (WEEE) COM(2008)810/4
COUNCIL OF EUROPEAN MUNICIPALITIES AND REGIONS CONSEIL DES COMMUNES ET REGIONS D EUROPE Registered in the Register of Interest Representatives of the European Commission. Registration number: 81142561702-61
More informationProbabilistic Worst-Case Response-Time Analysis for the Controller Area Network
Probabilistic Worst-Case Response-Time Analysis for the Controller Area Network Thomas Nolte, Hans Hansson, and Christer Norström Mälardalen Real-Time Research Centre Department of Computer Engineering
More informationON WEIGHTED RECTANGLE PACKING WITH LARGE RESOURCES*
ON WEIGHTED RECTANGLE PACKING WITH LARGE RESOURCES* Aleksei V. Fishkin, 1 Olga Gerber, 1 Klaus Jansen 1 1 University of Kiel Olshausenstr. 40, 24118 Kiel, Germany {avf,oge,kj}@informatik.uni-kiel.de Abstract
More informationOn the Maximum Throughput of A Single Chain Wireless Multi-Hop Path
On the Maximum Throughput of A Single Chain Wireless Multi-Hop Path Guoqiang Mao, Lixiang Xiong, and Xiaoyuan Ta School of Electrical and Information Engineering The University of Sydney NSW 2006, Australia
More informationDecentralizing the Semantic Web: Who will pay to realize it?
Decentralizing the Semantic Web: Who will pay to realize it? Tobias Grubenmann, Daniele Dell Aglio, Abraham Bernstein, Dmitry Moor, Sven Seuken Department of Informatics, University of Zurich, Switzerland,
More informationThe Pennsylvania State University. The Graduate School. Department of Computer Science and Engineering
The Pennsylvania State University The Graduate School Department of Computer Science and Engineering QUALITY OF SERVICE BENEFITS OF FLASH IN A STORAGE SYSTEM A Thesis in Computer Science and Engineering
More informationSome problems in ad hoc wireless networking. Balaji Prabhakar
Some problems in ad hoc wireless networking Balaji Prabhakar Background Example scenarios for ad hoc packet networks - sensor networks (many nodes, low data rates) - wireless LANs (fewer nodes, high data
More informationIntroduction. Linear because it requires linear functions. Programming as synonymous of planning.
LINEAR PROGRAMMING Introduction Development of linear programming was among the most important scientific advances of mid-20th cent. Most common type of applications: allocate limited resources to competing
More informationMigration Based Page Caching Algorithm for a Hybrid Main Memory of DRAM and PRAM
Migration Based Page Caching Algorithm for a Hybrid Main Memory of DRAM and PRAM Hyunchul Seok Daejeon, Korea hcseok@core.kaist.ac.kr Youngwoo Park Daejeon, Korea ywpark@core.kaist.ac.kr Kyu Ho Park Deajeon,
More informationSIMULATION-BASED COMPARISON OF SCHEDULING TECHNIQUES IN MULTIPROGRAMMING OPERATING SYSTEMS ON SINGLE AND MULTI-CORE PROCESSORS *
SIMULATION-BASED COMPARISON OF SCHEDULING TECHNIQUES IN MULTIPROGRAMMING OPERATING SYSTEMS ON SINGLE AND MULTI-CORE PROCESSORS * Hala ElAarag, David Bauschlicher, and Steven Bauschlicher Department of
More informationUsing Architectural Models at Runtime: Research Challenges
Proceedings of the European Workshop on Software Architectures, St. Andrews, Scotland, May 2004. Using Architectural Models at Runtime: Research Challenges David Garlan and Bradley Schmerl Department of
More informationOverview Computer Networking What is QoS? Queuing discipline and scheduling. Traffic Enforcement. Integrated services
Overview 15-441 15-441 Computer Networking 15-641 Lecture 19 Queue Management and Quality of Service Peter Steenkiste Fall 2016 www.cs.cmu.edu/~prs/15-441-f16 What is QoS? Queuing discipline and scheduling
More informationAdvanced Operations Research Prof. G. Srinivasan Department of Management Studies Indian Institute of Technology, Madras
Advanced Operations Research Prof. G. Srinivasan Department of Management Studies Indian Institute of Technology, Madras Lecture 16 Cutting Plane Algorithm We shall continue the discussion on integer programming,
More informationDISTRIBUTED NETWORK RESOURCE ALLOCATION WITH INTEGER CONSTRAINTS. Yujiao Cheng, Houfeng Huang, Gang Wu, Qing Ling
DISTRIBUTED NETWORK RESOURCE ALLOCATION WITH INTEGER CONSTRAINTS Yuao Cheng, Houfeng Huang, Gang Wu, Qing Ling Department of Automation, University of Science and Technology of China, Hefei, China ABSTRACT
More information5 QUESTIONS TO ASK WHEN CONSIDERING SERVER SIDE HEADER BIDDING
5 QUESTIONS TO ASK WHEN CONSIDERING SERVER SIDE HEADER BIDDING Introduction Server side (s2s) integrations have existed in programmatic advertising for nearly a decade, having been the preferred method
More informationLinear Programming. L.W. Dasanayake Department of Economics University of Kelaniya
Linear Programming L.W. Dasanayake Department of Economics University of Kelaniya Linear programming (LP) LP is one of Management Science techniques that can be used to solve resource allocation problem
More informationforward packets do not forward packets
Power-aware Routing in Wireless Packet Networks Javier Gomez, Andrew T. Campbell Dept. of Electrical Engineering Columbia University, N 10027, USA Mahmoud Naghshineh, Chatschik Bisdikian IBM T.J. Watson
More informationThe Fly & Anti-Fly Missile
The Fly & Anti-Fly Missile Rick Tilley Florida State University (USA) rt05c@my.fsu.edu Abstract Linear Regression with Gradient Descent are used in many machine learning applications. The algorithms are
More informationThe Index Coding Problem: A Game-Theoretical Perspective
The Index Coding Problem: A Game-Theoretical Perspective Yu-Pin Hsu, I-Hong Hou, and Alex Sprintson Department of Electrical and Computer Engineering Texas A&M University {yupinhsu, ihou, spalex}@tamu.edu
More information1 Dynamic Programming
CS161 Lecture 13 Dynamic Programming and Greedy Algorithms Scribe by: Eric Huang Date: May 13, 2015 1 Dynamic Programming The idea of dynamic programming is to have a table of solutions of subproblems
More informationFIGURE 3. Two-Level Internet Address Structure. FIGURE 4. Principle Classful IP Address Formats
Classful IP Addressing When IP was first standardized in September 1981, the specification required that each system attached to an IP-based Internet be assigned a unique, 32-bit Internet address value.
More informationFrom Software Sensitivity to Software Dynamics: Performance Metrics for Real-Time Software Architectures
From Software Sensitivity to Software Dynamics: Performance Metrics for Real-Time Software Architectures Janusz Zalewski Computer Science Department Florida Gulf Coast University Ft. Myers, FL 33965-6565,
More informationAccelerometer Gesture Recognition
Accelerometer Gesture Recognition Michael Xie xie@cs.stanford.edu David Pan napdivad@stanford.edu December 12, 2014 Abstract Our goal is to make gesture-based input for smartphones and smartwatches accurate
More informationModule 1 Lecture Notes 2. Optimization Problem and Model Formulation
Optimization Methods: Introduction and Basic concepts 1 Module 1 Lecture Notes 2 Optimization Problem and Model Formulation Introduction In the previous lecture we studied the evolution of optimization
More informationGraphical Analysis. Figure 1. Copyright c 1997 by Awi Federgruen. All rights reserved.
Graphical Analysis For problems with 2 variables, we can represent each solution as a point in the plane. The Shelby Shelving model (see the readings book or pp.68-69 of the text) is repeated below for
More informationDesigning Views to Answer Queries under Set, Bag,and BagSet Semantics
Designing Views to Answer Queries under Set, Bag,and BagSet Semantics Rada Chirkova Department of Computer Science, North Carolina State University Raleigh, NC 27695-7535 chirkova@csc.ncsu.edu Foto Afrati
More information3.1 Graphing Using the First Derivative
3.1 Graphing Using the First Derivative Hartfield MATH 2040 Unit 3 Page 1 Recall that a derivative indicates the instantaneous rate of change at an x-value (or other input value) which is related to the
More informationOPTIMIZING A VIDEO PREPROCESSOR FOR OCR. MR IBM Systems Dev Rochester, elopment Division Minnesota
OPTIMIZING A VIDEO PREPROCESSOR FOR OCR MR IBM Systems Dev Rochester, elopment Division Minnesota Summary This paper describes how optimal video preprocessor performance can be achieved using a software
More informationCS334: Digital Imaging and Multimedia Edges and Contours. Ahmed Elgammal Dept. of Computer Science Rutgers University
CS334: Digital Imaging and Multimedia Edges and Contours Ahmed Elgammal Dept. of Computer Science Rutgers University Outlines What makes an edge? Gradient-based edge detection Edge Operators From Edges
More informationVisualization and Analysis of Inverse Kinematics Algorithms Using Performance Metric Maps
Visualization and Analysis of Inverse Kinematics Algorithms Using Performance Metric Maps Oliver Cardwell, Ramakrishnan Mukundan Department of Computer Science and Software Engineering University of Canterbury
More informationFOUNDATIONS OF A CROSS-DISCIPLINARY PEDAGOGY FOR BIG DATA *
FOUNDATIONS OF A CROSS-DISCIPLINARY PEDAGOGY FOR BIG DATA * Joshua Eckroth Stetson University DeLand, Florida 386-740-2519 jeckroth@stetson.edu ABSTRACT The increasing awareness of big data is transforming
More informationA Graph-based Approach to Compute Multiple Paths in Mobile Ad Hoc Networks
A Graph-based Approach to Compute Multiple Paths in Mobile Ad Hoc Networks Gunyoung Koh, Duyoung Oh 1 and Heekyoung Woo 2 1 School of Electrical Engineering and Computer Science Seoul National University,
More informationIN5050: Programming heterogeneous multi-core processors Thinking Parallel
IN5050: Programming heterogeneous multi-core processors Thinking Parallel 28/8-2018 Designing and Building Parallel Programs Ian Foster s framework proposal develop intuition as to what constitutes a good
More informationEdge and local feature detection - 2. Importance of edge detection in computer vision
Edge and local feature detection Gradient based edge detection Edge detection by function fitting Second derivative edge detectors Edge linking and the construction of the chain graph Edge and local feature
More informationObjectives. Materials
Activity 13 Objectives Understand what a slope field represents in terms of Create a slope field for a given differential equation Materials TI-84 Plus / TI-83 Plus Graph paper Introduction One of the
More informationCONSUMPTION BASICS. MICROECONOMICS Principles and Analysis Frank Cowell. July 2017 Frank Cowell: Consumption Basics 1
CONSUMPTION BASICS MICROECONOMICS Principles and Analysis Frank Cowell July 2017 Frank Cowell: Consumption Basics 1 Overview Consumption: Basics The setting The environment for the basic consumer optimisation
More informationSUGGESTED SOLUTION CA FINAL MAY 2017 EXAM
SUGGESTED SOLUTION CA FINAL MAY 2017 EXAM ADVANCED MANAGEMENT ACCOUNTING Test Code - F M J 4 0 1 6 BRANCH - (MULTIPLE) (Date : 11.02.2017) Head Office : Shraddha, 3 rd Floor, Near Chinai College, Andheri
More information6LPXODWLRQÃRIÃWKHÃ&RPPXQLFDWLRQÃ7LPHÃIRUÃDÃ6SDFH7LPH $GDSWLYHÃ3URFHVVLQJÃ$OJRULWKPÃRQÃDÃ3DUDOOHOÃ(PEHGGHG 6\VWHP
LPXODWLRQÃRIÃWKHÃ&RPPXQLFDWLRQÃLPHÃIRUÃDÃSDFHLPH $GDSWLYHÃURFHVVLQJÃ$OJRULWKPÃRQÃDÃDUDOOHOÃ(PEHGGHG \VWHP Jack M. West and John K. Antonio Department of Computer Science, P.O. Box, Texas Tech University,
More informationContextion: A Framework for Developing Context-Aware Mobile Applications
Contextion: A Framework for Developing Context-Aware Mobile Applications Elizabeth Williams, Jeff Gray Department of Computer Science, University of Alabama eawilliams2@crimson.ua.edu, gray@cs.ua.edu Abstract
More informationHow to Exploit Abstract User Interfaces in MARIA
How to Exploit Abstract User Interfaces in MARIA Fabio Paternò, Carmen Santoro, Lucio Davide Spano CNR-ISTI, HIIS Laboratory Via Moruzzi 1, 56124 Pisa, Italy {fabio.paterno, carmen.santoro, lucio.davide.spano}@isti.cnr.it
More informationA Study on Issues Associated with Mobile Network
Available Online at www.ijcsmc.com International Journal of Computer Science and Mobile Computing A Monthly Journal of Computer Science and Information Technology IJCSMC, Vol. 3, Issue. 9, September 2014,
More informationIEEE networking projects
IEEE 2018-18 networking projects An Enhanced Available Bandwidth Estimation technique for an End-to-End Network Path. This paper presents a unique probing scheme, a rate adjustment algorithm, and a modified
More information5. DUAL LP, SOLUTION INTERPRETATION, AND POST-OPTIMALITY
5. DUAL LP, SOLUTION INTERPRETATION, AND POST-OPTIMALITY 5.1 DUALITY Associated with every linear programming problem (the primal) is another linear programming problem called its dual. If the primal involves
More informationSpatial Transform Technique
The overlooked advantage of the two-pass spatial transform technique is that complete, continuous, and time-constrained intensity interpolation can be conveniently achieved in one dimension. A Nonaliasing,
More informationFeedback-based dynamic proportion allocation for disk I
Oregon Health & Science University OHSU Digital Commons CSETech December 1998 Feedback-based dynamic proportion allocation for disk I Dan Revel Dylan McNamee Calton Pu David Steere Jonathan Walpole Follow
More informationSemantic Web in a Constrained Environment
Semantic Web in a Constrained Environment Laurens Rietveld and Stefan Schlobach Department of Computer Science, VU University Amsterdam, The Netherlands {laurens.rietveld,k.s.schlobach}@vu.nl Abstract.
More informationInteractive Campaign Planning for Marketing Analysts
Interactive Campaign Planning for Marketing Analysts Fan Du University of Maryland College Park, MD, USA fan@cs.umd.edu Sana Malik Adobe Research San Jose, CA, USA sana.malik@adobe.com Eunyee Koh Adobe
More informationChapter 7. Conclusions and Future Work
Chapter 7 Conclusions and Future Work In this dissertation, we have presented a new way of analyzing a basic building block in computer graphics rendering algorithms the computational interaction between
More informationEnd-To-End Delay Optimization in Wireless Sensor Network (WSN)
Shweta K. Kanhere 1, Mahesh Goudar 2, Vijay M. Wadhai 3 1,2 Dept. of Electronics Engineering Maharashtra Academy of Engineering, Alandi (D), Pune, India 3 MITCOE Pune, India E-mail: shweta.kanhere@gmail.com,
More informationIn other words, we want to find the domain points that yield the maximum or minimum values (extrema) of the function.
1 The Lagrange multipliers is a mathematical method for performing constrained optimization of differentiable functions. Recall unconstrained optimization of differentiable functions, in which we want
More informationWorkloads Programmierung Paralleler und Verteilter Systeme (PPV)
Workloads Programmierung Paralleler und Verteilter Systeme (PPV) Sommer 2015 Frank Feinbube, M.Sc., Felix Eberhardt, M.Sc., Prof. Dr. Andreas Polze Workloads 2 Hardware / software execution environment
More informationA Search Theoretical Approach to P2P Networks: Analysis of Learning
A Search Theoretical Approach to P2P Networks: Analysis of Learning Nazif Cihan Taş Dept. of Computer Science University of Maryland College Park, MD 2742 Email: ctas@cs.umd.edu Bedri Kâmil Onur Taş Dept.
More informationParallel Algorithms for the Third Extension of the Sieve of Eratosthenes. Todd A. Whittaker Ohio State University
Parallel Algorithms for the Third Extension of the Sieve of Eratosthenes Todd A. Whittaker Ohio State University whittake@cis.ohio-state.edu Kathy J. Liszka The University of Akron liszka@computer.org
More informationA Short SVM (Support Vector Machine) Tutorial
A Short SVM (Support Vector Machine) Tutorial j.p.lewis CGIT Lab / IMSC U. Southern California version 0.zz dec 004 This tutorial assumes you are familiar with linear algebra and equality-constrained optimization/lagrange
More informationTopic: Local Search: Max-Cut, Facility Location Date: 2/13/2007
CS880: Approximations Algorithms Scribe: Chi Man Liu Lecturer: Shuchi Chawla Topic: Local Search: Max-Cut, Facility Location Date: 2/3/2007 In previous lectures we saw how dynamic programming could be
More informationOptimal Configuration of Compute Nodes for Synthetic Aperture Radar Processing
Optimal Configuration of Compute Nodes for Synthetic Aperture Radar Processing Jeffrey T. Muehring and John K. Antonio Deptartment of Computer Science, P.O. Box 43104, Texas Tech University, Lubbock, TX
More informationFinite Element Method. Chapter 7. Practical considerations in FEM modeling
Finite Element Method Chapter 7 Practical considerations in FEM modeling Finite Element Modeling General Consideration The following are some of the difficult tasks (or decisions) that face the engineer
More informationQ: Which month has the lowest sale? Answer: Q:There are three consecutive months for which sale grow. What are they? Answer: Q: Which month
Lecture 1 Q: Which month has the lowest sale? Q:There are three consecutive months for which sale grow. What are they? Q: Which month experienced the biggest drop in sale? Q: Just above November there
More informationCSMA based Medium Access Control for Wireless Sensor Network
CSMA based Medium Access Control for Wireless Sensor Network H. Hoang, Halmstad University Abstract Wireless sensor networks bring many challenges on implementation of Medium Access Control protocols because
More informationLeveraging Transitive Relations for Crowdsourced Joins*
Leveraging Transitive Relations for Crowdsourced Joins* Jiannan Wang #, Guoliang Li #, Tim Kraska, Michael J. Franklin, Jianhua Feng # # Department of Computer Science, Tsinghua University, Brown University,
More information7 Fractions. Number Sense and Numeration Measurement Geometry and Spatial Sense Patterning and Algebra Data Management and Probability
7 Fractions GRADE 7 FRACTIONS continue to develop proficiency by using fractions in mental strategies and in selecting and justifying use; develop proficiency in adding and subtracting simple fractions;
More informationCompositional Schedulability Analysis of Hierarchical Real-Time Systems
Compositional Schedulability Analysis of Hierarchical Real-Time Systems Arvind Easwaran, Insup Lee, Insik Shin, and Oleg Sokolsky Department of Computer and Information Science University of Pennsylvania,
More informationData Warehousing Alternatives for Mobile Environments
Data Warehousing Alternatives for Mobile Environments I. Stanoi D. Agrawal A. El Abbadi Department of Computer Science University of California Santa Barbara, CA 93106 S. H. Phatak B. R. Badrinath Department
More informationFormal Model. Figure 1: The target concept T is a subset of the concept S = [0, 1]. The search agent needs to search S for a point in T.
Although this paper analyzes shaping with respect to its benefits on search problems, the reader should recognize that shaping is often intimately related to reinforcement learning. The objective in reinforcement
More information